Instructions to use HuggingFaceH4/starchat2-15b-v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use HuggingFaceH4/starchat2-15b-v0.1 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("text-generation", model="HuggingFaceH4/starchat2-15b-v0.1") messages = [ {"role": "user", "content": "Who are you?"}, ] pipe(messages)# Load model directly from transformers import AutoTokenizer, AutoModelForCausalLM tokenizer = AutoTokenizer.from_pretrained("HuggingFaceH4/starchat2-15b-v0.1") model = AutoModelForCausalLM.from_pretrained("HuggingFaceH4/starchat2-15b-v0.1") messages = [ {"role": "user", "content": "Who are you?"}, ] inputs = tokenizer.apply_chat_template( messages, add_generation_prompt=True, tokenize=True, return_dict=True, return_tensors="pt", ).to(model.device) outputs = model.generate(**inputs, max_new_tokens=40) print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:])) - Notebooks
- Google Colab
- Kaggle
- Local Apps
- vLLM
How to use HuggingFaceH4/starchat2-15b-v0.1 with vLLM:
Install from pip and serve model
# Install vLLM from pip: pip install vllm # Start the vLLM server: vllm serve "HuggingFaceH4/starchat2-15b-v0.1" # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:8000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "HuggingFaceH4/starchat2-15b-v0.1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker
docker model run hf.co/HuggingFaceH4/starchat2-15b-v0.1
- SGLang
How to use HuggingFaceH4/starchat2-15b-v0.1 with SGLang:
Install from pip and serve model
# Install SGLang from pip: pip install sglang # Start the SGLang server: python3 -m sglang.launch_server \ --model-path "HuggingFaceH4/starchat2-15b-v0.1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "HuggingFaceH4/starchat2-15b-v0.1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }'Use Docker images
docker run --gpus all \ --shm-size 32g \ -p 30000:30000 \ -v ~/.cache/huggingface:/root/.cache/huggingface \ --env "HF_TOKEN=<secret>" \ --ipc=host \ lmsysorg/sglang:latest \ python3 -m sglang.launch_server \ --model-path "HuggingFaceH4/starchat2-15b-v0.1" \ --host 0.0.0.0 \ --port 30000 # Call the server using curl (OpenAI-compatible API): curl -X POST "http://localhost:30000/v1/chat/completions" \ -H "Content-Type: application/json" \ --data '{ "model": "HuggingFaceH4/starchat2-15b-v0.1", "messages": [ { "role": "user", "content": "What is the capital of France?" } ] }' - Docker Model Runner
How to use HuggingFaceH4/starchat2-15b-v0.1 with Docker Model Runner:
docker model run hf.co/HuggingFaceH4/starchat2-15b-v0.1
the model loves to generate: // Your code here
Hi, thanks for training the model!
Have you noticed that the model really likes to complete code with:
// Your code here
or its python version of
# Your code here
I try to run some evals on it, and in many cases, it just completes my code with this placeholder.
I know it is a chat model, and it works excellent for conversions, but I wanted to take it for humaneval spin to check its programming abilities, and I get for python much less than your 70 because it doesn't want to generate my the code in many cases :(
any idea?
It works good for code or text completion.
Here is my system prompt:
You are a text completion machine.
Finish the following text starting from the last symbol:
Test case:
In JavaScript, the .find() function is used on arrays to locate the first element in the array that satisfies the provided testing function.
In Python, a similar functionality can be achieved using a list comprehension or the filter() function.
Here's how you can achieve the equivalent of .find() in JavaScript using Python:
Using list comprehen
Answer:
sion:
def find_in_python(lst, func):
for element in lst:
if func(element):
return element
return None # If no matching element is found
# model answer....
....